A pixel unit design method for an optical encoder

Abstract

The present invention provides a pixel unit design method for an optical encoder. The receiving unit (2) used in the design is composed of rectangular pixel units (4) and does not need to be processed into a special shape. The pixel unit (4) used is generally a photodiode (PD). Depending on the application, a photoelectric detection device such as an avalanche photodiode (APD) and a single photon avalanche diode (SPAD) can also be selected. The receiving unit (2) needs to cooperate with a scanning light spot (3) generated by other optical systems to generate a periodic signal. When the device is working, it can be regarded as the scanning light spot (3) moving along the scanning direction (1) and passing through the receiving unit (2) to generate a sine-like periodic signal. The present invention can be used for receiving optical signals in optical encoder products and can also be used for measuring optical signals in grating ruler products.

Description

Technical Field

[0001] The present invention relates to a pixel unit design method for an optical encoder, which can be used for optical encoder and grating ruler measurement and belongs to the field of photoelectric detection technology. Background Art

[0002] A photoelectric encoder is a sensor that converts the mechanical geometric displacement of an output shaft into pulses or digital values through photoelectric conversion. This is the most widely used sensor. Using photoelectric conversion technology, a photoelectric encoder converts the geometric displacement of a rotating shaft into pulses of equal width for output. This discretizes the continuous displacement into individual pulses of equal width. The generated pulses correspond one-to-one with the displacement. Therefore, the smaller the displacement corresponding to a pulse, the more accurate the measurement. The sum of the recorded pulses corresponds to the total displacement. Typically, an incremental photoelectric encoder outputs two pulse signals 90° out of phase and generates a Z pulse signal for each rotation. By analyzing the phase relationship between the pulse signals, the direction of rotation can be easily determined; the Z phase can be used to reduce cumulative error. Therefore, the receiving pixel is an indispensable component of a photoelectric encoder product.

[0003] In the field of photoelectric detection technology, photodetectors commonly used in receiving pixels include photodiodes, avalanche photodiodes, and single-photon avalanche diodes. With the rise of semiconductor production and integrated circuit design in this century, the manufacturing technology for these devices has advanced significantly. Compared to mainstream photoelectric sensors from the last century, such as photomultiplier tubes, these devices offer numerous advantages, including small size, light weight, excellent thermal stability, and ease of use. Photodiodes have also been widely used in photoelectric encoder products. To generate a sinusoidal signal, the photocurrent generated by the photodiode requires system-level decoding, or the photodiode must be finely segmented. Advanced signal processing is then applied to the signals generated by these diodes to provide users with high-resolution periodic signals. However, in integrated circuits, the price of ADCs increases rapidly with increasing resolution, impacting manufacturing costs.

[0004] In order to reduce the dependence on post-filtering circuits, researchers at home and abroad have studied and designed the shape of photodiodes or other types of detectors in sensor pixels under existing production processes to optimize the performance and cost of the overall product. In 2015, Xiong Wenshuo, Xiong Wenkai and others published a scheme for the coordinated design of a graticule pattern, a code disk pattern and a pixel array (CN 204831337), in which the graticule is designed to have slits in multiple long-axis directions, the code disk uses multiple code channels in accordance with the Gray code design scheme, and the pixel array uses multiple pixel units, with the size and arrangement in the spot scanning direction and the vertical scanning direction modulated to improve the system accuracy. Since the center part of the pixel array is not continuous, the graticule needs to be precisely aligned with the pixel array during installation, which is inconvenient in production and manufacturing; iC-Haus GmbH published a pixel shape design scheme for an optical encoder (DE 10 2014 112 459 Optischer Position sensor) designs a metal mask with a special opening shape for the arrow-shaped pixel unit, so that the scanning spot formed by the LED and the code disk generates a half-cycle sinusoidal signal when passing through the pixel unit. Through subsequent circuit processing, multiple signals such as sine and cosine can be obtained. However, metal mask processing of the pixel unit increases process costs on the one hand, and scratches randomly introduced when processing the pixel unit surface on the other hand will reduce the product yield. Dr. Johannes Heidenhain GmbH also proposed a pixel unit shape design scheme (DE 0541829 Vorrichtung zur Erzeugung oberwellenfreier periodisher Signale). When the scanning spot passes through a single pixel, a half-cycle sinusoidal signal can be generated. The disadvantage is that directly cutting the pixel into advanced geometric shapes will cause a series of process problems such as edge collapse, which will damage the pixel itself.

[0005] To address the above issues, the present invention discloses a method for designing receiving pixels for a linear optical encoder that generates a quasi-sinusoidal waveform signal. This method can be used in optical encoders, linear scale measurement, and other fields. The present invention modulates the pixel array in the vertical scanning direction to generate a quasi-sinusoidal waveform signal. Its structure and process are simple: the pixels are all rectangular, avoiding the cost, yield, and process challenges associated with using complex patterns to cut pixels or metal masks to cover them. Summary of the Invention

[0006] The present invention aims to provide a pixel unit design method for an optical encoder that generates a quasi-sinusoidal waveform signal. The main factors considered in this design method include:

[0007] 1. A scanning spot (3) that can move or scan along a scanning direction (1);

[0008] 2. The scanning spot (3) can be generated by the light source and the code disk on the plane where the receiving unit (2) is located, and is characterized in that the light power density in the scanning spot area is consistent;

[0009] 3. A receiving unit (2), composed of a plurality of pixel units (4), capable of receiving light signals from the scanning light spot and converting the light signals into electrical signals for output;

[0010] 4. The pixel unit (4) can be a photodiode, an avalanche photodiode, or a single-photon avalanche diode. The pixel unit (4) is divided into two different areas in the receiving unit (2): a forward signal area (5) and a reverse signal area (6), and each pixel unit in each area has a different length but the same width.

[0011] The present invention realizes its function according to the following principles:

[0012] When the scanning spot (3) moves along the scanning direction (1), it first passes through the forward signal area (5) in the receiving unit (2). The area where the scanning spot (3) overlaps with each pixel unit (4) will generate an electrical signal. Since the increment of the overlapping area is not completely linear with the increase of the scanning stroke, when the scanning spot (3) scans the three pixel units (7), (8) and (9) in the forward signal area (5), a half-cycle quasi-sinusoidal signal is generated; when the scanning spot (3) continues to scan until it reaches the center of the two areas, the signal intensity returns to zero; when the scanning spot (3) continues to move forward, the reverse signal area (6) will continue to generate the second half of the quasi-sinusoidal cycle signal. In the process of the scanning spot (3) moving forward, only the pixel units in the overlapping area will contribute to the generated signal.

[0013] The scanning direction (1) in the system can be either a linear direction or a rotational direction centered on a certain point.

[0014] The receiving unit (2) in the system is composed of a plurality of pixel units (4) having the same size in the direction of motion but different sizes in the orthogonal direction, which are symmetrically distributed. The size occupied by these pixel units in the direction of motion is P, and the size occupied by a single pixel unit is P / N, where N is the number of pixel units in the receiving unit (2). In theory, the more pixel units there are, the higher the signal accuracy.

[0015] The scanning spot (3) in the system is generated by other optical systems on the plane where the pixel unit (4) is located. Its size in the moving direction is 2 pixels to ensure that when it moves to the P / 2 position, it just covers two or more pixel units at the center position of the receiving unit (2), ensuring that the entire receiving unit (2) outputs a "0" signal at this time.

[0016] Preferably, the sizes of the pixels (7)(8)(9) in the vertical direction are not equal, and the sizes of these pixels are modulated.

[0017] The pixel unit (4) in the system is usually composed of a photodiode. Preferably, an avalanche photodiode, a single photon avalanche diode or other photoelectric sensor device can also be selected. Its production process includes but is not limited to modern semiconductor processes such as complementary metal oxide semiconductor (CMOS), and its morphology includes but is not limited to rectangular, willow leaf, fan-shaped, etc.

[0018] When the width of the scanning spot (2) is 2 pixel units (4), the size ratio of each pixel perpendicular to the scanning direction along the scanning direction should follow the following formula:

[0019]

[0020] Among them L n is the relative length of the nth pixel unit, N is the number of pixel units in a receiving unit, where N is usually an even number and L0=0. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 The present invention is a schematic diagram of the system structure of the receiving end of an optical encoder for generating a quasi-sinusoidal waveform signal. The receiving end system mainly consists of a scanning light spot (3) and a receiving unit (2), wherein the receiving unit (2) is composed of a plurality of pixel units (4) of equal size in the scanning direction (1) but not equal size in the vertical direction. During operation, the light in the scanning light spot (3) is uniform, and the receiving unit (2) is scanned along the scanning direction (1). Since the pixel units (4) have a specific shape distribution design, the system can generate a quasi-sinusoidal waveform optical signal. The width of the scanning light spot (3) can be adjusted according to the system accuracy. The width of the scanning light spot (3) shown in the figure is the width of two pixel units (4).

[0022] Figure 2This is a detailed structural diagram of the receiving unit (2). The receiving unit (2) is mainly composed of a plurality of pixel units (4); these pixel units (4) are further divided into two groups, one group is the forward signal area (5) responsible for outputting the forward signal, and the other group is responsible for outputting the reverse signal, called the reverse signal area (6). When the scanning spot (3) scans along the scanning direction (1), it first scans the pixel unit (7) that generates the forward signal and starts to generate the forward signal; when scanning to the middle of the forward signal area (5) and the reverse signal area (6), since the two pixel units (9) and (10) covered by the scanning spot have equal areas and receive equal light energy, two signals of equal size and opposite directions are generated to cancel each other out, thereby outputting a signal "0" with a phase of 180°. When the scanning spot (3) continues to scan and enters the reverse signal area (6), the receiving unit (2) starts to output the reverse signal, corresponding to the signal in the second half of the sine cycle.

[0023] Figure 3 The present invention is a schematic diagram of a first embodiment of a method for designing a pixel unit of an optical encoder for generating a quasi-sinusoidal waveform signal. In this design example, the default width of the scanning spot (3) is 2 pixel units, the number of pixel units (4) in the receiving unit (2) is N=6, and scanning starts from the pixel unit (4) numbered a; in order to achieve the output of the quasi-sinusoidal waveform signal, the length of each pixel unit (4) needs to be modulated, and the length ratio of each numbered pixel unit is shown in the diagram.

[0024] Figure 4 for Figure 3 The implementation effect of the first embodiment is supplemented, and the figure shows the quasi-sine waveform signal generated by the receiving unit.

[0025] Figure 5 The present invention is a schematic diagram of a second embodiment of a method for designing a pixel unit of an optical encoder for generating a quasi-sinusoidal waveform signal. In this design example, the default width of the scanning spot (3) is 2 pixel units, and the number of pixel units (4) in the receiving unit (2) is N=10. The scanning spot (2) scans along the scanning direction (1). In order to achieve the output of the quasi-sinusoidal waveform signal, the length ratio of each pixel unit (4) needs to be modulated.

[0026] Figure 6 for Figure 5 The implementation effect of the second embodiment is supplemented, and the figure shows the quasi-sine waveform signal generated by the receiving unit. DETAILED DESCRIPTION

[0027] Example 1:

[0028] Figure 3This paper presents an implementation example of a pixel unit design method for an optical encoder that generates a sinusoidal signal. The receiving unit consists of six photodiodes with the same dimensions in the scanning direction but different dimensions in the vertical scanning direction. Because the receiving unit size must be modulated to generate a sinusoidal signal, the following formula is used to calculate the photodiode size:

[0029]

[0030] By calculation, the size ratio of each photodiode in the vertical scanning direction can be obtained: a = 0.707106, b = 0.292893, c = 0.414213, d = 0.414213, e = 0.292893, f = 0.707106. In this embodiment, the size of the photodiode along the scanning direction is 10um, and the maximum size perpendicular to the scanning direction can be 100um. By proportion, the size of each photodiode in the vertical scanning direction is a = 70.7106um, b = 29.2893um, c = 41.4213um, d = 41.4213um, e = 29.2893um, f = 70.7106um.When the scanning spot (2) scans the photodiode a, the overlapping area between them increases linearly with the scanning motion, and its slope is ka=70.7106 / D; wherein D is the width of a single photodiode along the scanning direction, in this embodiment, D=10um, then ka=7.07106; when the scanning spot (2) scans the photodiode b, the overlapping area between them increases linearly with the scanning motion, while the overlapping area with the photodiode a remains unchanged, then the generated signal changes with the scanning motion with a slope kb=2.92893; when the scanning spot (2) scans the photodiode c, the overlapping area between them increases linearly with the scanning motion, but the scanning spot (2) and the photodiode c remain unchanged. The overlapping area between the photodiodes a decreases linearly, and the generated signal changes with the scanning motion with a slope of kc=41.4213 / D-70.7106 / D. It can be calculated that kc=-2.92893; when the scanning spot (2) sweeps over the photodiode d, the overlapping area between them increases linearly with the scanning motion, but the overlapping area with the photodiode b decreases linearly. Taking into account the reverse signal output by the photodiode d, the generated signal changes with the scanning motion with a slope of kd=-41.4213 / D-29.28983 / D=-7.07106; when the scanning spot (2) sweeps over the photodiode e, the overlapping area between them increases linearly with the scanning motion, but the overlapping area with the photodiode b decreases linearly. The overlapping area between the photodiodes c decreases linearly with the scanning motion, and the generated signal changes with the scanning motion at a slope of ke=-29.4213 / D-41.4213 / D=-7.07106; when the scanning spot (2) scans the photodiode f, the overlapping area between them increases linearly with the scanning motion, but the overlapping area with the photodiode d decreases linearly with the scanning motion, and the generated signal changes with the scanning motion at a slope of kf=-70.7106 / D-(-41.4213 / D)=-2.92893; when the scanning spot (2) scans the photodiode f and continues to move forward along the scanning direction for a pixel unit width D, the scanning spot (2) The area of overlap with the photodiode f remains unchanged with the scanning motion, but the area of overlap with the photodiode e decreases linearly with the scanning motion, so the slope of the generated signal changes with the scanning motion is kf'=29.2893 / D=2.92893; when the scanning spot (2) scans a distance of D~2D behind the photodiode f, the area of overlap between the photodiode f and the scanning spot (2) decreases linearly with the scanning motion, but there is no photodiode in the space behind the photodiode f to receive the signal, so the slope of the generated signal changes with the scanning motion is kf"=70.7106 / D=7.07106. The overall change of the signal with the scanning motion of the scanning spot (2) is as follows. Figure 4As shown, the overall generation of a sine-like waveform signal is achieved. In addition, it can also be shaped and filtered in conjunction with the back-end circuit to make the overall waveform more consistent with the sine waveform, or deformed into a cosine waveform, etc.

[0031] Example 2:

[0032] Figure 5 This paper presents an implementation example of a pixel unit design method for an optical encoder that generates a sinusoidal signal. The receiving unit consists of 10 photodiodes with the same dimensions in the scanning direction but different dimensions in the vertical scanning direction. Because the receiving unit size must be modulated to generate a sinusoidal signal, the following formula is used to calculate the photodiode size:

[0033]

[0034] By calculation, the size ratio of each photodiode in the vertical scanning direction can be obtained as follows: a = 0.5, b = 0.366025, c = 0.633975, d = 0.232050, e = 0.267950, f = 0.267950, g = 0.232050, h = 0.633975, i = 0.366025, j = 0.5. In this embodiment, the size of the photodiode along the scanning direction is 10 μm. The maximum size perpendicular to the scanning direction can be 100um. The sizes of each photodiode perpendicular to the scanning direction are calculated by proportion: a=50um, b=36.6025um, c=63.3um, d=41.4213um, e=26.7950um, f=26.7950um, g=23.2050um, h=63.3975um, i=36.6025um, j=50um. When the scanning spot (2) scans the photodiode a, the overlapping area between them increases linearly with the scanning motion, and its slope is ka=50 / D; wherein D is the width of a single photodiode along the scanning direction. In this embodiment, D=10um, then ka=5; when the scanning spot (2) scans the photodiode b, the overlapping area between them increases linearly with the scanning motion, while the overlapping area with the photodiode a remains unchanged, then the generated signal changes with the scanning motion with a slope kb=3.66025; when the scanning spot (2) scans the photodiode c, the overlapping area between them increases linearly with the scanning motion, but the overlapping area between the scanning spot (2) and the photodiode a decreases linearly, then the generated signal changes with the scanning motion with a slope kc=63.3975 / D-50 / D, and it can be calculated that kc=1.33975; when the scanning spot (2) scans the photodiode d, the overlapping area between them increases linearly with the scanning motion, while the overlapping area between the scanning spot (2) and the photodiode a decreases linearly, then the generated signal changes with the scanning motion with a slope kc=63.3975 / D-50 / D, and it can be calculated that kc=1.33975; The overlapping area increases linearly with the scanning motion, but the overlapping area with photodiode b decreases linearly, so the generated signal changes with the scanning motion with a slope of kd = 23.2050 / D-36.6025 / D = -13.3975; when the scanning spot (2) scans the photodiode e, the overlapping area between them increases linearly with the scanning motion, but the overlapping area with photodiode c decreases linearly with the scanning motion, so the generated signal changes with the scanning motion with a slope of ke = 26.7950 / D-63.3975 / D = -3.66025; when the scanning spot (2) scans the photodiode f, the overlapping area between them increases linearly with the scanning motion, but the overlapping area with photodiode d decreases linearly with the scanning motion. Taking into account the reverse signal output by photodiode f, the generated signal changes with the scanning motion with a slope of kf = -26.7950 / D-23.2050 / D=-5;When the scanning spot (2) passes through the photodiode g, the overlapping area between them increases linearly with the scanning motion, but the overlapping area between them and the photodiode e decreases linearly with the scanning motion. Taking into account the reverse signal output by the photodiode g, the slope of the generated signal with the scanning motion is kg=-23.205 / D-26.7950 / D=-5;When the scanning spot (2) passes through the photodiode h, the overlapping area between them increases linearly with the scanning motion, but the overlapping area between them and the photodiode f decreases linearly with the scanning motion. Considering that the photodiode h outputs a reverse signal, the slope of the generated signal with the scanning motion is kh=-63.3975 / D-(-26.7950 / D)=-3.66025; when the scanning spot (2) scans the photodiode i, the overlapping area between them increases linearly with the scanning motion, but the overlapping area with the photodiode g decreases linearly with the scanning motion. Considering that the photodiode i outputs a reverse signal, the slope of the generated signal with the scanning motion is ki=-36.6025 / D-(-26.7950 / D)=-1.3397 5; When the scanning light spot (2) scans the photodiode j, the overlapping area between them increases linearly with the scanning motion, but the overlapping area with the photodiode h decreases linearly with the scanning motion. Taking into account the reverse signal output by the photodiode j, the slope of the generated signal with the scanning motion is kj=-50 / D-(-63.3975 / D)=1.33975; When the scanning light spot (2) scans the photodiode j and continues to move forward along the scanning direction by a pixel unit width D, the overlapping area between the scanning light spot (2) and the photodiode j remains unchanged with the scanning motion. However, the overlapping area with the photodiode i decreases linearly with the scanning motion, so the generated signal changes with the scanning motion at a slope kj'=36.6025 / D=3.66025; when the scanning spot (2) scans a distance D to 2D behind the photodiode j, the overlapping area between the photodiode j and the scanning spot (2) decreases linearly with the scanning motion, but there is no photodiode in the space behind the photodiode j to receive the signal, so the generated signal changes with the scanning motion at a slope kj"=50 / D=5. The overall change of the signal with the scanning motion of the scanning spot (2) is as follows. Figure 6 As shown, the overall generation of a sine-like waveform signal is achieved. In addition, it can also be shaped and filtered in conjunction with the back-end circuit to make the overall waveform more consistent with the sine waveform, or deformed into a cosine waveform, etc.

Claims

1. A method for designing a pixel unit of an optical encoder, characterized by: The receiving end of the optical encoder is composed of a scanning light spot (3) moving along a scanning direction (1) and a receiving unit (2); the scanning light spot (3) in the system is generated by an optical system on the plane where the receiving unit (2) is located, and has a uniform light power density distribution inside; when the scanning light spot (3) moves along the scanning direction (1), it passes through the receiving unit (2); the receiving unit (2) is composed of a plurality of pixel units (4), which can receive light signals and convert the received light power into current or voltage signals for output; when the scanning light spot (3) is projected onto the pixel unit (4), a signal is generated that is proportional to the overlapping area of the scanning light spot (3) and the pixel unit (4); the sizes of these pixel units (4) in the scanning direction are equal, and in the direction perpendicular to the scanning direction, their sizes can be modulated according to the waveform of the sinusoidal signal, thereby outputting a signal similar to a sinusoidal waveform; The size of each pixel unit (4) constituting the receiving unit (2) in the vertical scanning direction needs to be modulated. Along the scanning direction (1), the distribution law of the size complies with: in Indicates the size ratio of the nth pixel unit in the vertical scanning direction, N represents the number of pixel units (4) in the receiving unit (2), N is an even number and After solving the ratio of each pixel unit (4) through this proportional relationship, a size parameter A can be determined by combining it with the actual production process, multiplied by To determine the size of a single pixel unit in the system; in actual production and manufacturing, A is any one of 100um, 50um, 20um, and 10um; when the scanning spot (3) scans the receiving unit (2), it overlaps with the pixel units (4) of different sizes, generating signals; these signals are superimposed on each other, and finally generate a sine-like waveform signal output.

2. The method for designing a pixel unit of an optical encoder according to claim 1, wherein: It is necessary to consider the scanning spot, which is generated by an LED light source in combination with a collimating lens, or by a vertical cavity surface semiconductor laser array in combination with a light homogenizer or a microlens array; when the light beam generated by one of the methods passes through a structure with a rectangular opening, the generated light spot is a scanning spot with a uniform light power density distribution, and the structure with a rectangular opening is a code disk of an optical encoder; this scanning spot has a certain size in the scanning direction, with the size of a pixel unit (4) in the scanning direction as the basic unit, and the size of the scanning spot (3) is required to be twice the size of the pixel unit (4) in the scanning direction; the size of the scanning spot perpendicular to the scanning direction is required to be larger than any pixel unit (4), to ensure that each pixel unit has the opportunity to receive the light signal that fills the entire pixel unit during the scanning process.

3. The method for designing a pixel unit of an optical encoder according to claim 1, wherein: A receiving unit (2) is provided, the receiving unit being located in the scanning direction (1), or the scanning light spot (3) being located in the same plane and at the same horizontal position; the receiving unit (2) is composed of a plurality of pixel units (4), and the pixel units are arranged in contact with each other at the same horizontal position; These pixel units (4) are photoelectric detection devices, including photodiodes, avalanche photodiodes, or single-photon avalanche photodiodes; when these pixel units (4) constitute the receiving unit (2), they are rectangular, triangular, willow-leaf-shaped, or fan-shaped; these pixel units (4) are further subdivided into two areas: a forward signal area (5) and a reverse signal area (6), and the final output signal is the sum of the signals output by the pixel units in the two areas.

4. The method for designing a pixel unit of an optical encoder according to claim 2, wherein: According to actual measurement requirements, the opening on the code disk of the optical encoder is replaced with a fan-shaped or Z-shaped one, thereby changing the shape of the scanning spot (3); when the shape of the scanning spot (3) changes, the shape of the receiving unit (2) is adapted to the scanning spot (3) to ensure the accuracy of the generated signal, and the shape of the receiving unit (2) is changed to the same shape as the scanning spot (3) and reduced by a specific proportional coefficient; or the shape of the scanning spot (3) is split and reorganized into another shape, which is reduced by a specific proportional coefficient to serve as the shape of the receiving unit (2).

5. The method for designing a pixel unit of an optical encoder according to claim 3, wherein: The arrangement is adjusted so that, under the premise of ensuring that each pixel unit (4) has the opportunity to receive the light signal that fills the entire pixel unit (4), the position of the pixel unit (4) in the scanning direction (1) remains unchanged, and the pixel units (4) are staggered in a direction perpendicular to the scanning direction (1) to avoid contact between them.

Images